通过关联从 has many 中查找匹配和不匹配记录的单次出现

Question

我有 3 个 Postgres table

组
groups_favourites
最爱

这 3 个 table 的结构是：

groups
--------------------------
| id |  name   | user_id |
--------------------------
| 1  | Group A |   100   |
| 2  | Group B |   100   |
| 3  | Group C |   100   |
| 4  | Group D |   100   |
--------------------------

favourites_groups
-------------------------------
| id | group_id | favorite_id |
-------------------------------
| 1  |  2       | 10          |
| 2  |  2       | 12          |
| 3  |  2       | 14          |
| 4  |  2       | 15          |
| 5  |  2       | 16          |
| 6  |  3       | 12          |
| 7  |  4       | 10          |
| 8  |  4       | 11          |
| 9  |  4       | 12          |
| 10 |  4       | 13          |
-------------------------------
UNIQUE INDEX ON [group_id, favorite_id]

favourites
-------------------------------
| id | product_id  | user_id  |
-------------------------------
| 10 |    1000     |   100    |
| 11 |    1001     |   100    |
| 12 |    1002     |   100    |
| 13 |    1003     |   100    |
| 14 |    1004     |   100    |
| 15 |    1005     |   100    |
| 16 |    1006     |   100    |
-------------------------------

我在后端的 Rails 上使用 Ruby，这 3 个 table 之间的关系是通过 [=14= 设置的]协会

我想要完成的结果是：对于给定的产品和用户，我想找到所有在 favorite_groups table 中有条目的组 + 所有没有条目的组在 favorites_groups table.

假设 user_id = 100 AND product_id = 1000，结果将是：

| group_id | group_name | favorite_id | product_id | ---------------------------------------------------- | 1 | Group A | NULL | NULL | | 2 | Group B | 10 | 1000 | | 3 | Group C | NULL | NULL | | 4 | Group D | 10 | 1000 | ----------------------------------------------------

另一个用例是 user_id = 100 AND product_id = 1002，结果是：

| group_id | group_name | favorite_id | product_id | ---------------------------------------------------- | 1 | Group A | NULL | NULL | | 2 | Group B | 12 | 1002 | | 3 | Group C | 12 | 1002 | | 4 | Group D | 12 | 1002 | ----------------------------------------------------

我正在使用的当前 SQL 给我 1 行 NULL favorite_id 和 product_id 以及 1 行匹配的 favorite_id 和 product_id

这是我的 SQL:

SELECT DISTINCT ON (group_id, favorite_id, product_id) groups.id AS group_id, groups.name AS group_name, favorites.id AS favorite_id, favorites.product_id AS product_id FROM groups LEFT JOIN favorite_groups ON favorite_groups.group_id = groups.id LEFT JOIN favorites ON favorites.id = favorite_groups.favorite_id AND favorites.product_id = 1002 WHERE groups.user_id = 100

由于此查询返回的结果，我必须额外工作以遍历每一行并准备我想要的结果集。

有更好的方法吗？

在此先感谢您的帮助！

Answer 1

您的模型有些奇怪。 groups.name 与 groups 和 favourites 中出现的 user_id 之间的关系尚不清楚。 favourites_groups 上的 unique 约束应该使 favourites 中的 user_id 变得不必要，所以我添加了一个注释掉的 join 条件。

请试试这个查询，看看它是否return是您需要的：

select g.id as group_id, g.name as group_name
  from groups g
  left join favourites_groups fg
    on fg.group_id = g.id
  left join favourites f
    on f.id = fg.favorite_id
 --  and f.user_id = g.user_id 
 where g.user_id = 100
   and f.product_id = 1002
;

更新对于那个很抱歉。这应该 return 你想要什么：

select g.id as group_id, g.name as group_name,
       max(f.id) as favorite_id, 
       max(f.product_id) as product_id
  from groups g
  left join favourites_groups fg 
    on fg.group_id = g.id
  left join favourites f 
    on f.id = fg.favorite_id
   and f.product_id = 1000
 where g.user_id = 100
 group by g.id, g.name
 order by g.id;

通过关联从 has many 中查找匹配和不匹配记录的单次出现

Find single occurrence of matched and non-matched records from has many through association

ruby-on-rails

postgresql

has-many-through